Breaking Down the Break Downs

Hello again everyone. I’m back after a week to recuperate and reflect. Seeing as my Double Masters Power Rankings will be live today on ChannelFireball, I wanted to do something a little different today. I want to talk about the metrics I use to examine Pauper metagame health.

My metagame analysis focuses on the Top 32 decklists from each Challenge. I work from this set for a few reasons. First, it is the only officially published data from Wizards of the Coast for these Challenges. Second, I want to see what consistently rises to the top. While crowdsourced matchup data is useful, I feel that looking at the decks that finish outside the Top 32 does little to inform what wins. Instead, it does a great job of telling me what the best decks are beating…but I can glean that information from the Top 32 results as well.

Does this approach cut me off from seeing part of the metagame? Absolutely! But when I look at Pauper I want to explore what’s doing well and doing so consistently. With that out of the way, let’s look at the headers.

Let’s get the self explanatory columns out of the way. Archetype is just that – the archetype for a specific deck. Sometimes I adjust these midseason but usually between seasons. For example, the Boros Monarch decks that ran a heavier black component (Unearth, Chainer’s Edict) are found under a different header – Mardu Monarch. Moving forward, this is not likely to be the case.

Volume measures the number of times a deck appears in the Top 32, while Top 8s and Wins measure exactly what they describe.

Win+: This metric is used to measure how well a deck does against the field in a given tournament. The metric is set to measure wins against the lowest finishing “positive” record deck (for the purposes of this measure, a .500 record is positive). In a 6 round event, a 3-3 record nets 0 points, a 4-2 nets 1, and so on to 6-0. In a 7 round event, 4-3 is the baseline (0). Win+ has proven to be a relatively accurate measure in the gross sense with decks that accumulate more over the course of a season being the decks that are seen as “top tier”. But where it really comes in handy is looking at Win+:Vol.

The ratio of Win+ score to the number of times a deck appears in the Top 32 is informative. The closer this ratio is to 1, the closer it is to averaging a Top 16 finish. If an archetype consistently is above one, season after season, you can infer the deck is either dominant or has very few entrants in the Top 32, all of which have done well. For example, Flicker Tron’s ratio is 0.92 for Double Masters season. Azorius Familiars has a 1.2 ratio. Yet Flicker Tron was over 10 times as popular as Familiars. Which deck is better? Which would you place higher on your Power Rankings?

Expected Top 8/%: This is one of my more controversial metrics. It takes the ratio and multiplies it by .3125. Why this number? On average, 2.5 decks with a Win+ of 1 will make the Top 8 of a given Pauper Challenge. Multiplying the ratio by this coefficient gives us the number of times we could expect a deck to make Top 8. This is measured against Real Top 8/% to determine if a deck is over or under performing in terms of Top 8 appearances (Delta %, Delta Top 8). What do these numbers tell us? They help to correct for when decks fall prey to variance. Sometimes a deck is quote unquote good and fails to make Top 8 in a given season. That doesn’t mean the deck is bad but maybe just fell on the wrong side of the coin for a given stretch. Over a long enough stretch, however, it helps to indicate trends in relative strength.

Let’s use Izzet Faeries as an example. In the post-ban Core 2021 season, the archetype performed 1.13 Top 8s above expectation. That number was nearly quartered in Ikoria season to 0.31 Top 8s above expectation (18.75% above to 0.63% above). Isolated, this could just be a bad run. Taken in context of Double Masters season (1.42% below expectations, 0.31 Top 8s below) let’s us know that something else is going on and warrants further exploration.

Volume % is also self explanatory, telling us how much space an archetype takes up in the metagame. Weighted Volume looks at the space occupied by that archetype’s Win+ score. The Delta (Volume) looks at the discrepancy between how popular a deck is and how well it performs. A positive delta indicates a deck is outperforming expectations, negative the converse. Let’s take Stompy for example. It was the second most popular archetype during Double Masters with 44 appearances – 11.46% of all Top 32 decks. It took down 9.39% of the total Win+ for the season (-2.07%). Again, as alone datum it means nothing. Looking at Ikoria, it had a positive skew of 0.13%. Is this enough to tell us anything? Not yet, but if this skid continues with Zendikar Rising then we can start to draw conclusions.

This is how I examine the Pauper winner’s metagame. It is far from comprehensive but it does provide a look at the relative strength of the top decks. It also is pretty good at spotting problematic decks, say ones that take up nearly a third of all available Top 8 slots.

If you like this sort of analysis, please consider signing up for my Patreon. If you have suggestions on ways to make my data more complete (using the officially published data), please let me know; I’m always looking for more ways to share the information that is out there